Unipolar depression is a disabling emotional response characterized by symptoms such as:depressed mood; loss of interest in pleasurable activities; fatigue; appetite and sleep disturbances;difficulties in thinking and concentrating; and repeated death or suicidal ideation (American
Psychiatric Association, 1994). As the Cross-National Collaborative Group (1992) concluded from a study involving over 40,000 subjects across the world, depression is on the rise. Presently, about 30% of all American adults have been depressed at some point in their lives and about 5% of all American adults are depressed at any given time (Richelson, 1993; as cited in Julien, 1996). Of the people who have been depressed, about 25% of them will eventually commit suicide (Richelson, 1993; as cited in Julien, 1996). On an economic level, the annual cost of depression and its concomitants (such as therapy, medication and occupational absenteeism) runs close to 43 billion dollars a year in the United States alone (American Psychiatric Association, 1997). Thus, depression is not only a debilitating and potentially fatal condition but also a costly one. Managing this "Depression Epidemic" has become a critical concern.
The purpose of this paper is to put forth a model of depression that can not only account for the demographics of depression but also for other important elements of the depression puzzle. However, before such an inclusive theory is proposed, the current understanding of depression will be described and the inadequacies of this set of beliefs will be discussed. Following this the interrelationship between stress and depression will be clarified. After presenting this proposed model, established patterns in depression such as: its apparent heritability, its increasing rates (through cohort, period and age of onset effects), gender differences in its occurrence, the prevalence of dual diagnoses, and the effectiveness of antidepressants will be explained. In the proposed model, neurochemical activity plays a small and not sufficient role. However, consistent with the current zeitgeist, considerable attention will be given to neurochemical issues (mostly regarding serotonin).
The current understanding of depression is premised upon the asserted assumptions that depression is a disease and that the apparent correlation between neurotransmitter system deficits and depression is causal in nature. There are several outgrowths from this position, most notably the biological amine theory (depression results from a functional deficit of specific neurotransmitters at certain sites in the brain) and the hypersensitive postsynaptic receptor theory (depression is not a function of low levels of neurotransmitters per se, but rather because of the consequent development of hyperresponsive postsynaptic receptors - Julien, 1996). Nevertheless, the dominant theory of our times is the serotonin hypothesis which conforms to the hypersensitive postsynaptic receptor theory with the additional claim that serotonin plays the biggest neurochemical role in depression (Julien, 1996). Unfortunately, the classical serotonin hypothesis attributes any defects in the serotonergic system to genetic factors and, in essence, reduces depression solely into its genetically determined biochemical constituents. Although the causal link between serotonin and depression has been hypothesized for at least the last 25 years, only very recently has there been any visual evidence of defective serotonergic transmission (Mann & Malone, 1996). However, mere visual evidence of serotonergic malfunction still leaves many issues unresolved. Most notably, it is absurd to consider depression as mainly a function of a single defective neurotransmitter system as neurotransmitters do not act in isolation. Serotonin not only functions in other somatic processes outside the brain, but serves to modulate other neurotransmitter substances in the brain. Thus, shortages of serotonin can lead to wild fluctuations in the activity of other neurotransmitters (Davison & Neale, 1996; Ashby Jr., 1996.). Therefore, other neurotransmitters almost surely play a neurochemical role in depression.
Most of the supporting evidence for the serotonin theory has been obtained through drug action studies. However, the success of drugs that act directly on the serotonergic system (or any other neurotransmitter system) and can alleviate depressive symptoms is not proof that depression is entirely or dominantly biological. Such an assertion is, at best, reverse logic (Breggin, 1994). For example, if relief for a fever is obtained after taking an aspirin, this does not tell us anything about the cause of the fever. The mechanisms by which aspirin may alleviate the symptoms is not necessarily related to the etiology of the fever (food poisoning, a flu virus, drug withdrawal etc.). Similarly, just because Prozac may alleviate depressive symptoms (through increasing the amount of serotonin available to the postsynaptic receptors) does not in any way demonstrate that a lack of serotonin caused the depression. Hence, it is highly possible that drugs may influence the affective symptomology of depression without acting on the true etiology. Therefore, it is very problematic to rely on pharmacological response for concluding the etiology of depression. As a further note, even the efficacy of antidepressants is still under question by some investigators as up to 90% of depressed subjects in some studies improve on placebo alone (Greenberg & Fisher, 1989; as cited in Breggin, 1994). Considering that depression can often lift on its own, the actual efficacy of antidepressants may be partly inflated due to natural remissions. Also in disconcordance with the neurotransmitter theories, manually depleting the brain of neurotransmitters (a reverse action to that of antidepressants) should then cause depression; and it does, but only in about 20% of subjects (all of whom had either already been depressed in the past or were undergoing considerable life stress at the time - Goodwin & Bunney Jr, 1971; as cited in Breggin, 1994). Though this may be taken as partial support, most other studies that induced the biochemical abnormalities associated with depression have not resulted in the disorder in any of the subjects (Lesher, 1982). Considering this, the neurochemical abnormality associated with depression appears to be a necessary but not sufficient factor in depression.
As mentioned, the serotonin hypothesis likens depression to a physiological disease. However, if depression is truly a disease then mourning the loss of a loved one for longer than two weeks (the DSM-IV length of depressive symptoms in order to meet the criteria for Major Depressive Disorder) is a physiological disease. Further, the notion of depression as purely organic is not consistent with the fact that many depressed individuals resolve their depression without any drug intervention whatsoever. Therefore, the current focus on neurotransmitters is incorrect. A much better conceptualization would see the grieving process as part of an expected human condition that undoubtably takes people different lengths of time to understand and thereby manage the psychosocial consequences of such a loss. In other words, depression is best conceptualized as a response to painful or stressful life circumstances and reflects a difficulty in adapting to life stress. Considering this, depression can be seen as desisting when the person regains adequate levels of coping resources and a feeling of self-efficacy (as opposed to when neurotransmitter levels are corrected). Therefore, any view which reduces depression purely into its biochemical constituents (as the neurotransmitter theories do) and ignores how depression functions in and interacts with its context and environment does the construct of depression much injustice. While it is true that depression has a neurochemical basis, the fact of the matter is that anything human (from thumbsucking to thinking to feeling depressed) involves neurochemical activity; that is just the way the human body is constructed. The biochemical changes associated with depression are better seen as necessary symptoms of depression and not the defining elements of the disorder. There is evidence to support this notion as simple instructions to continually think sad thoughts can lead to dysphoric mood and the neurochemical concomitants of depression (Teasdale & Bancroft, 1977; as cited in Beck & Harrison, 1982). Similarly, merely repeatedly rehearsing negative self-statements can also lead to the same neurochemical changes (May & Johnston, 1973; as cited in Beck & Harrison, 1982).
The futility of the serotonin hypothesis of depression is further exemplified in it's inability to explain many "pieces" of the depression puzzle. For example, while it can account for the apparent genetic heritability of depression (inherit a faulty serotonergic system), the serotonin hypothesis can not account for gender differences in depression. While women are at least twice as likely as men to become depressed (Nolen-Hoeksema, 1990; as cited in Davison & Neal, 1996) there is no evidence that women are born with a less effective serotonergic system (Watson, 1997). As well, the serotonin hypothesis can only explain the high comorbidity of depression with other psychiatric disorders by calling into awareness that all must involve serotonin (which does not tell us very much because almost every function of the body involves serotonin to a certain degree). Furthermore and most notably, the serotonin hypothesis can not adequately account for the gradual increase in depression across the world. According to the serotonin hypothesis, the world can only become more depressed if there is an increase in defective serotonergic systems throughout the world; and there is no evidence of this (Watson, 1997). Similarly, the fact that people are becoming depressed at younger ages can only be explained by the serotonin hypothesis if serotonergic systems are becoming defective at younger ages. However, there is also no evidence for this (Watson, 1997). Furthermore, it is also counter-evolutionary that a defective serotonergic system will be continually passed on through "the survival of the fittest" especially when depressed people are more likely to commit suicide (and not pass on their genes). Therefore, the serotonin hypothesis and the neurotransmitter theories, as they stand, prove fruitless in explaining the depression puzzle.
A better avenue to explore in explaining the depression puzzle is the relationship between chronic stress and depression. A stressor can be defined as any stimulus that threatens an individual's physical, psychological or social health or disturbs the his or her homeostasis. Stress is the sum of the physical, behavioural and emotional reactions to any of these threatening stimuli and can be classified into eustress (stress resulting from pleasant stressors) and distress (stress resulting from unpleasant situations - Insel, Roth, Rollins, & Peterson, 1996). Important for this discussion, chronic stress refers to the condition resulting from the continued presence of one or more distressful stressors. As much research has shown (see the review by Anisman & Zacharko, 1982) stressful experiences result in both behavioural attempts to cope with the stressor(s) as well as an increased utilization of serotonin, norepinephrine and dopamine (as part of the physiological stress response) to contend with the environmental demands as well as maintain the biological functioning of the individual (Anisman & Zacharko, 1992). However, if distress is experienced over a prolonged period, neurotransmitter utilization may exceed synthesis and result in a lower overall level of neurotransmitters. Furthermore, after resolving the chronic stressor(s), exposure to a previously mild stressor that ordinarily has little stressful effect, can result in further neurotransmitter depletion (Anisman & Sklar, 1979; as cited in Anisman & Zacharko, 1982, Anisman & Zacharko, 1992). The use of animal models to bridge the gap between chronic stress and depression has proved fruitful. Aprilson and Hintgen's (1981; as cited in Aprilson & Hintgen, 1993) 5-HTP animal model of depression extends the basic serotonin hypothesis as it recognizes that serotonergic deficits may not only be due to genetic factors but also be the result of stress. Other animal models also have bridged the gap between stress and depression (Willner, Moreau, Nielsen, Papp & Sluzewaska, 1996). For example, in rats and mice, chronic exposure to distressful stressors decreases hedonic responsiveness, escape performance, appetitive response, sexual behaviour and body weight and also results in sleep disturbances - symptoms quite similar to human depression (Aprilson & Hingtgen, 1992; Cheeta, Boekkamp & Willner, 1996; as cited in Willner et al., 1996). Further, these effects are effectively reversed by treatment with electroconvulsive therapy and appropriate doses of antidepressants and not with amphetamines, neuroleptics and other non-antidepressants (Willer et al, 1996). Taken together, the available data support the chronic stress animal model of depression and it appears to possess face, predictive and construct validity.
Though most of the above mentioned research has been conducted in animal studies, further support can be found through studies of humans as well. Depression is associated with a higher frequency of antecedent stressful life events (Anisman & Zacharko, 1982). Further, stressful life events (as measured by the Holmes and Rahe (1967) scale) have shown predictive value in later onset of depression (Hanskell & Prusoff, 1975; as cited in Anisman & Zacharko, 1982). Similarly, the more unpleasant (and usually chronic) life events of the Holmes and Rahe (1967) scale were the ones most strongly related to later depression (Lewinsohn & Amenson, 1978; as cited in Anisman & Zacharko, 1982). As well, early life trauma (such as parental death) that undoubtably result in chronic stress (parental absence has broad-ranging implications) increases vulnerability to later depression (Lloyd, 1980a; as cited in Anisman & Zacharko, 1982). Consistent evidence can also be obtained from examining the period fluctuations of depression in Beirut. Beirut experienced a dramatic increase in depression between 1950 and 1960: a time of chaotic and stressful political and demographic change (Cross-national Collaborative Group, 1992). However, in the next decade, corresponding with economic prosperity and political stability (presumably a less stressful time) depression rates dropped sharply only to increase again over the next decade, concurrent with the highly stress-inducing Lebanese wars (National Collaborative Group, 1992).
It would appear that there is an important relationship between chronic stress and depression. As such, the stress-depression link will be formalized by proposing that failure to adapt to stress is the cause of depression. It should be noted that under this model, stress (or a stressor), in itself, does not cause depression. Rather, depression is more properly conceived of as a function of possessing inadequate resources for managing one's stress which results in a demanding increase in biochemical coping (adaptation) and less available neurotransmitters for other uses. In turn, this contributes to the physiological symptoms associated with depression (a schematic representation of this model is given in appendix A).
In elaboration of the model, an individual who becomes depressed is first faced with a chronic negative stressor, an acute stressor with long lasting or broad effects or a set of overlapping repeated minor stressors. The person's perceptions and attributions then transform the objective stressor(s) into a subjective level of experienced stress. If the individual has a negative cognitive structure, he or she may overexaggerate the importance of negative events and experience more subjective stress as a function of a specific stressor (Lloyd, 1980a, as cited in Beck and Harrison, 1982). In essence, perception and attribution together function like a mirror that subjectively reflects the experienced severity of the stressor (in terms of how stress-inducing it is). In response to the experienced stress, the person engages in psychological, behavioural and social coping strategies (such as exercise, time management strategies, confiding in a counsellor or trusted friend or vacationing). Specific to social resources, there is evidence that depressed persons have weaker social networks before becoming depressed and reduced social support lessens the ability of the person to effectively cope with life stress (Davison & Neale, 1996). When coping resources (which are also affected by attributions, perceptions and cognitive biases) are exhausted, an individual's last coping mechanism (neurochemical activity - the stress response) intensifies to maintain the biological functioning of the organism. After a prolonged state in which the neurochemical adaptation is intensified, loss of adequate neurochemical adaptation occurs (Selye, 1976; as cited in Insel, Roth, Rollins, & Peterson, 1996). Nonetheless, the biologically-related symptoms of depression (such as sleep and concentration disturbances) are a function of a lack of available neurotransmitters to contend with the stressor and provide for all the other functions that neurotransmitters serve. Returning to the dominant serotonin theory, serotonergic abnormalities are necessary but not sufficient to cause depression. Even if a person inherits a defective neurotransmitter system, if his or her attributions and perceptions are self-serving and he or she possesses a lot of coping resources, he or she will not likely become depressed. Similarly, depression is rarely caused by inducing the appropriate biochemical abnormalities because most individuals likely possess adequate resources to cope with their presently experienced life stress (or else they would have already become depressed). In further support of the model, Cushing's Syndrome, which involves a chronic oversecretion of cortisol, is physiologically analogous to experiencing an endless array of chronic stressors and hence should often lead to depression (as it does).
As can be inferred from the model, genetic heritability plays neither a necessary nor sufficient role in depression. Firstly, to clarify, a biological diathesis in one or more of the neurotransmitter systems can be passed on from parent to offspring. However, such a diathesis is not necessary for an individual to become depressed. Some examples of potential predisposers include presynaptic deficiencies such as: a defect in precursor transport properties, biosynthesis, neurotransmitter storage, autoreceptors, reuptake or catabolism; or postsynaptic deficiencies such as: a defect in receptor sensitivity, cycle nucleotides, phosphaditylinositol cycle products, or other second messenger system components (Aprilson & Hintgen, 1993). Also, because the neurochemical adaptation to stressors also involves dopamine-beta-hydroxlase, tyrosine hydroxylase and tryptophan hydroxylase, a genetically inherited deficiency in any of these enzyme systems can also result in insufficient or inefficient neurotransmitter breakdown. Congruently, children of depressed parents are at much greater risk for developing depression, other relatives of depressed people are more likely to be or become depressed and depression concordance rates for identical twins have consistently been found to be 2-3 times greater than for fraternal twins. As it appears, the family undoubtably plays a role in determining depression, but this role is not necessarily genetic. Although frequently claimed, there is no sound justification for assuming that familial aggregation of depression is entirely, mostly, or even partially genetic. The fact that parent and child are both depressed may not indicate a genetic component, but rather a social learning effect since there are many shared family experiences and considerable opportunities to observe and influence one another (especially in regards to coping strategies). Therefore, depression may appear to run in families because, on average, there is greater contact between a depressed individual and his or her twin than between the same individual and other immediate family members (and similarly, on average, there is also greater contact with immediate family members than with more distant family members). Another reason for the high heritability ascribed to depression and a critique of many adoption studies regards the fact that adoptive parents are frequently relatives, friends or strangers who similar to the family (race, parental age etc.) and hence either genetically, psychologically or socially similar (Kendler, 1993). This presents a definite bias that may result in an overestimation of the heritability of depression. Further, the traditional twin method is predicated on the equal environment assumption (that identical and fraternal twins each share the same similarity in environmental experiences). However, the physical similarity of identical twins may elicit more similar treatment by others as their childhood and adult environments are more similar than for fraternal twins (Kendler, 1993). As such, part of the genetically described greater concordance amongst identical twins is partly a function of environmental experience and hence it may be possible that the genetic element of depression is overestimated. Further, as research by Kendler (1995) has shown, the preferred practice of ascertaining probands for family studies through treatment facilities provides an overestimate of the familial risk of depression as treatment seeking is greater in those who already possess a family member with depression. Further, he concludes that this artifact accounts for about 16% to 30% of the increased risk of depression in family members.
In addition to the genetic aspects, it is important to examine the psychosocial consequences of growing up in a household with a depressed parent. The literature strongly suggests that depressed parents are markedly impaired in fulfilling their roles as parents and this too can account for some of the ascribed heritability of depression. Aside from the social learning effect, families in which a parent suffers from depression are more likely to have aversive interactions and chronic parental dysfunction, which undoubtably contribute to a negative outcome in the children (Marton & Maharaj, 1993). There is evidence that mothers suffering from depression are more critical of and less positive with their children than non-depressed mothers are (Hammen, 1991; as cited in Marton & Maharaj, 1993). Also in support, depressed parents are less likely to promote competence in their children (Marton & Maharaj, 1993).
Any proposed model of depression must be able to account for the increasing prevalence of depression as well as for its decreasing age of onset. The reason that more people are depressed now and are becoming so at decreasingly younger ages is because people are now under more stress than they can cope with. Exposure to a greater number of stressors (especially chronic ones) makes it more likely that an individual will not possess enough resources to deal with the experienced stress. Though there are countless possibilities for why the world is presently more stress-inducing, one reason for the rise may be a result of the increase in chronic medical disorders such as cardiovascular disease, AIDS and cancer (Insel, Roth, Rollins, & Peterson, 1996). As well, there is much greater exposure to carcinogens in the present time contributing to the regressively younger onset of these chronic diseases (Brown, 1996). The body's adaptation to these chronic diseases results in tremendous strain on part of the immune system (including neurochemical adaptation) to manage these lifelong diseases. Another possible reason for the increasing stress of the world results from the increasing sense of time urgency, as we are regularly trying to do more things in less time (Brown, 1996). The increasing perception that we do not have enough time to accomplish our goals results in regular psychological disappointments and provides another form of increasing chronic stress. There is also a present-day decrease in coping resources partly resulting from the less active lifestyle and unhealthier diets of present day society as well as the increase in single parent and teen parent families. In accounting for the decreasing age of onset, younger and younger individuals are facing the stress related to chronic diseases, time pressures and other distressful stimuli.
Any newly proposed model of depression must also be able to account for the increasing prevalence of depression amongst women. The reason for this gender difference relates to the fact that women are confronted with a greater number of stressors than men. In most parts of our gendered world, women still face a lack of personal, social and political power (Baker & Dryden, 1993; Davison & Neale, 1996). Based on their subordinate status, women usually endure much more chronic stress in the forms of sexism (especially in the workplace), domestic abuse (in which women are almost always the victims), greater pressure to conform idealistic beauty standards, greater incidence of sexual harassment, and greater rates of sexual trauma. Further, in many parts of the world, women are socialized into accepting a double duty role (work outside the home and inside the home) as well as other social roles that do not encourage them to feel competent (Bernard, 1973; as cited in Davison & Neale, 1996). Also, consistent with higher rates of depression amongst the poor (Hirschfeld & Cross, 1982; as cited in Davison & Neale, 1996), more women than men live in poverty and thus undergo more financial stressors. Aside from being subject to a higher level of chronic stressors, there is evidence that women also tend to perceive or attribute stimuli in a less self-serving manner. As Dweck (1975; as cited in Davison & Neale, 1997) has shown, women are more likely to attribute their success to luck or favours from others and their failure to global and stable factors, resulting in greater subjective stress than many men would experience. Further, women may be socialized to have less effective psychological and behavioural coping strategies which results in less ability to successfully adapt to stressors. For example, women, when coping with stress or sad feelings do not tend to engage in distracting activities (such as playing sports or watching television) as men tend to do, but rather ruminate in their distressing and sad feelings (Nolen-Hoeksema, 1990; as cited in Davison & Neale, 1996). This ruminating reaction may amplify the state of stress-related sadness, possibility interfering with personal coping strategies and utilization of other coping resources (Nolen-Hoeksema & Morrow, 1991b; as cited in Davison and Neale, 1996). Hence, women are more depressed than men because they face more chronic stressors and tend to possess less effective behavioural and psychological coping strategies.
Any model of depression should further be able to account for the undisputably high comorbity of depression with other disorders. Depression can either be the result of a chronic disorder, can worsen other conditions by preventing adaptation to further stressors or can spawn other disorders in order to help cope with the depression. Depression is most frequently comorbid with anxiety disorders (such as acute stress disorder, generalized anxiety disorder, agoraphobia, and social phobia) all of which can result in long lasting levels of anxiety (a form of chronic stress). In support, as longitudinal studies reveal, anxiety diagnoses typically precede depression (Rohde, Lewinsohn, and Seeley, 1991; as cited in Davison & Neale, 1996). Congruently, there is also a high comorbidity with other chronic conditions such as chronic pain disorder (Davis & Neale, 1997). Secondly, depression can also exacerbate other symptoms or hinder the ability (biochemical, social, behavioural and psychological) to deal with further stressors which can contribute to another psychiatric condition. This might explain the comorbity with less chronic psychiatric and physiological conditions. Thirdly, depression can often result in a substance disorder (or another disorder) to help deal with one's depressive symptoms as almost one in three depressed individuals also abuse alcohol or another drug (Kettl, 1992). The common reasons for using drugs (decreasing anxiety, reducing distress, inducing relaxation and promoting positive moods) appear to lend support to the notion that these disorders may develop for the purpose of dealing with one's depressive symptoms.
Because of the present zeitgeist, a distinct hindrance for any non-neurochemical theory of depression is the necessity to explain why antidepressants can work in treating depression. The answer can be inferred by understanding why some classes of antidepressants work better for certain subtypes of depression as well as for the observation that antidepressants are also useful in treating other psychiatric disorders. As stated by Anisman and Zacharko (1982), different subgroups of depression may be neurochemically distinct. Therefore, different classes of antidepressants may be better matched to certain neurochemical deficits. The reason for this supposition stems from the possibility that stress may differentially influence neurochemically activity across individuals. In addressing the effectiveness of antidepressants, if neurotransmitter utilization is increased (as is the case in chronic stress and depression), the individual may be less able to deal with further stressful experiences as well as provide for the many other functions of neurotransmitters (Anisman & Zacharko, 1992). The administration of antidepressants increases the amount and efficiency of certain neurotransmitters, affording more biochemical resources. On a less neurochemical level, the symptoms of depression (sleep and appetite disturbances, difficulties in thinking, fatigue etc.) definitely contribute to the depressed individual's inability to manage further stressors and other disorders may result. Merely alleviating the depression symptomology (in which several of the symptoms are not limited to depression) affords the individual more biochemical, social, behavioural and psychological resources to deal with the life stress and depression (or another serotonin/norepinephrine-related disorder) because the medicated individual not only possesses more biochemical resources, but has more energy, is likely more socially active and can sleep and think better.
Depression is on the rise necessitating a scholarly focus on understanding it in order to effectively prevent its occurrence. A model of depression based on chronic stress was proposed to provide a new framework for understanding depression. Chronic stress induces behavioural disturbances and neurochemical changes, many of which are reminiscent of the symptoms and neurochemical alterations associated with depression. Human studies, and more so animal studies, have provided much support for the validity of the model. Further, the model can successfully account for several demographics and other patterns of the depression problem, namely the reason for the increase in depression, the apparent genetic heritability, gender differences, and high comorbidity. Also, as demanded by the current zeitgeist, the reason for the success of the antidepressants in treating depression and other disorders was briefly addressed.
American Psychiatric Association (1997). Too few people are treated for depression. APA Monitor, 28 (3), 6.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, D.C.: APA.
Anisman, H. & Zacharko, R.M. (1982). Depression: the predisposing influence of stress. Behavioural and Brain Sciences, 5 (1), 89-137.
Anisman, H. & Zacharko, R.M. (1992). Depression as a consequence of inadequate neurochemical adaptation in response to stressors. British Journal of Psychiatry, 160, (suppl. 15), 36-43.
Aprilson, M.H. & Hintgen, J.N. (1993). A neurochemist's perspective on human depression and stress. In Kariya, T. & Nakagawa (1992) Perspectives on basic research and clinical practice.
Ashby Jr., C.R. (1996). The modulation of dopamine transmission by other neurotransmitters . New York: CRC.
Baker, M. & Dryden, J. (1993). Families in Canadian Society (2nd ed.). Toronto: McGraw-Hill Ryerson.
Beck, A.T. & Harrison, R.P. (1982). Stress, neurochemical substrates and depression: Concomitants are not necessarily causes. Peer commentary in Anisman, H. & Zacharko, R.M. (1982). Depression: the predisposing influence of stress. Behavioural and Brain Science, 5 (1), 101-102.
Breggin, P.R. (1994). Talking back to prozac. New York: St. Martin's Press.
Brown, S. (1997). Kinesiology 140 lecture. September 1996.
Cross-national Collaborative Group (1992). The changing rate of major depression. JAMA, 268 (21), 3098 - 3015.
Davison, G.C. & Neale, J.M. (1996). Abnormal psychology (6th ed.). New York: John Wiley & Sons.
Insel, P.M., Roth, W.T., Rollins, L.M. & Peterson, R.A. (1996). Core concepts in health. Toronto: Mayfield.
Julien, R.M. (1996). A primer of drug action (7th ed.). New York: W.H. Freeman.
Kendler, K.S. (1993). Twin studies of psychiatric illness: Current status and future directions. Archives of General Psychiatry, 50, 905-915.
Kendler, K.S. (1995). Is seeking treatment for depression predicted by a history of depression in relatives? Implications for family studies of affective disorder. Psychological Medicine, 25, 807-814.
Kettl, P. (1992). Alcoholism, drug abuse and depression. At Depression and related affective disorders association (DRADA) website: http://infonet.welch.jhu.edu/departments/drada/articles/articl03.htm.
Lesher, A.I. (1982). An alternate hypothesis of depression. Peer commentary for Anisman, H.& Zacharko, R.M. (1982). Depression: the predisposing influence of stress. Behavioural and Brain Sciences, 5 (1), 111-112.
Mann, J.J. & Malone, K.M. (1996). Demonstration of in vivo of reduced serotonin responsivity in the brain of untreated depressed patients. American Journal of Psychiatry, 152 (2), 174-182.
Marton, P. & Maharaj, M.A. (1993). Family Factors in Adolescent unipolar depression. Canadian Journal of Psychiatry, 38, 373-381.
Watson, N. (1997). Personal communication, Feb. 7th , 1997.
Willner, P., Moreaqu, J., Nielson, C.K., Papp, M. & Sluzewska, A. (1996). Decreased Hedonic responsiveness following chronic mild stress is not secondary to loss of body weight. Physiology and Behaviour, 60 (1), 129-134.